Chromatin structure and architecture. Histone proteins package and condense genomic DNA into chromatin. In its compact form, chromatin assumes a 30-nm fiber in vitro, whereas in its active form it assumes a 10-nm open structure. In vivo, chromatin is thought to adopt similar structures. Within the cell nucleus, proteins such as the CCCTC binding factor (CTCF) help direct chromatin higher-order organization through passive and active mechanisms by imposing topological constraints and actively participating in transcription events. CTCF is a highly conserved DNA binding protein found exclusively in bilaterians. The protein consists of an eleven zinc-finger DNA binding domain, flanked by conserved N-terminal and C-terminal domains that constitute about 57 percent of the protein. While the zinc-fingers specifically recognize DNA and interact with RNA, the roles of the N- and C-terminal domains are still unknown. To further our understanding of chromatin structure in vivo, and better understand the role of CTCF, we focused our studies on the domains flanking the DNA binding zinc fingers. We have previously shown that these N- and C-terminal domains are intrinsically disordered, a property that is shared by many related DNA binding proteins. We have characterized the physical nature of these unstructured and conserved domains. Current work focuses on the identification of high-affinity protein partners that bind to the N- and C-termini of CTCF and a study of the complexes formed, to understand how CTCF, and related proteins, regulate higher-order genome organization within the eukaryotic nucleus. Macromolecular assemblies of biological interest. Hydrodynamic methods, particularly sedimentation velocity and sedimentation equilibrium analytical centrifugation, are used to characterize critical biological assemblies and obtain information on their shape, stoichiometry, and interaction affinity. In collaboration with the Clore lab, we have focused our attention on the N-terminal fragment of the Huntingtin protein, encoded by the first exon of the HTT gene. This fragment contains a poly(Q) tract of variable length, typically of 17 to 20 glutamines. Repeat lengths greater than 36 invariably result in Huntingtins disease, thought to be controlled by the poly(Q) driven neurotoxic aggregation of the protein. The first self-association events of this fragment are challenging to study in vitro, as even non-pathogenic poly(Q) constructs aggregate rapidly. Reducing the length of the poly(Q) segment results in more soluble fragments and slows the kinetics of amyloid nucleation. Using a construct containing only seven glutamines we showed by sedimentation equilibrium that the fragment self-associates to form both dimers and tetramers. The dimerization affinity is rather weak with a dissociation constant of nine millimolar, whereas the subsequent tetramerization affinity is one thousand orders of magnitude stronger at eight micromolar, showing that dimer rapidly transforms into a tetramer. Using a series of mutants, we also showed that the stability of the dimer, relative to the monomer, drives the extent of tetramerization. Both sedimentation velocity and sedimentation equilibrium experiments were consistent with observations made by nuclear magnetic resonance (NMR) and electron paramagnetic resonance (EPR) that reported on the affinity and kinetics of these self-associations. NMR data also provided structural information and demonstrated how poly(Q) lengths greater than 36 can lead to efficient fibril nucleation.